Why does the formula lack the $ symbol after OCR, making it unable to render in the md editor?

[Fanxi1992]

📚 The doc issue

The formula for identifying academic papers is:
[ \begin{aligned} \vec{OD} &= (XI - XO, YI - YO) = \ &= \left( \frac{\sum_{i=1}^{n} (I_i)}{\sum_{i=1}^{n} P_i}, \frac{\sum_{i=1}^{n} (O_i)}{\sum_{i=1}^{n} P_i} \right) - \left( \frac{\sum_{i=1}^{n} (O_i)}{\sum_{i=1}^{n} P_i}, \frac{\sum_{i=1}^{n} (I_i)}{\sum_{i=1}^{n} P_i} \right) \end{aligned} ]

This cannot be recognized in editors like Typora.

If it is:

$[ \begin{aligned} \vec{OD} &= (XI - XO, YI - YO) = \ &= \left( \frac{\sum_{i=1}^{n} (I_i)}{\sum_{i=1}^{n} P_i}, \frac{\sum_{i=1}^{n} (O_i)}{\sum_{i=1}^{n} P_i} \right) - \left( \frac{\sum_{i=1}^{n} (O_i)}{\sum_{i=1}^{n} P_i}, \frac{\sum_{i=1}^{n} (I_i)}{\sum_{i=1}^{n} P_i} \right) \end{aligned} ]$

Then it can be identified, is this a negligence?

Suggest a potential alternative/fix

No response

[jakep-allenai]

Yeah, this is a limitation of V1 of this model. I think in a future version we are going to enforce a more consistent format on the equation syntax, and to try to maintain consistency from page to page as well.